PROJECT SUMMARY The emergence and transmission of antimicrobial resistance (AMR) is currently one of the most critical public health threats. As such there is an urgent need to identify new ways to reduce AMR transmission. In recent years, several studies have started to reveal the complex relations and interactions between antibiotics, antimicrobial resistance genes (ARGs) and the human microbiome. The human microbiome plays profound roles in responding to antibiotic treatment and the transmission of ARGs. The altered microbiome, especially in children, may have a distinct resistome (the entire ARGs in the microbiome community). Infants and young children receive frequent antimicrobial courses for common respiratory infections and are thought to spread ARGs effectively in the community. Otitis media is the most common reason for antimicrobial treatment in children. The antibiotic treatment that is selected for use in young children may significantly influence AMR spread in the communities. To date, there are a limited number of prospective, controlled comparative studies on how selected antibiotics affect the development of antibiotic resistance in the microbiome in young children. It is not possible to investigate children in an experimental way and expose them to antibiotics without an indication for the treatment. Non-severe acute otitis media, however, is an excellent clinical model to investigate the impact of different antimicrobial agents, since clinical guidelines suggest several alternative approaches to treat children, including watchful waiting without antibiotics. In this study, we will investigate the impact of the most commonly used antibiotics on the intestinal microbiome, and in particular the resistome in children with otitis media. The Oulu University study is IRB approved and has started enrolling infants and young children with non-severe otitis media in a study cohort and plan to finish the enrollment within the first few months of this project. Children are randomly allocated in four antibiotic treatment groups to compare the impact of amoxicillin, amoxicillin-clavulanate, azithromycin, or no treatment on the intestinal microbiome and ARGs. We will solicit the antibiotic exposure during lifetime and collect stool samples before (day 0) and three days after the antibiotic course (day 10). We will apply quantitative PCR, 16S rDNA and metagenomic sequencing on different platforms to characterize the microbiome and resistome over time. De novo genome assembly based on Oxford Nanopore long reads technology, combined with high depth Illumina sequencing will reveal the details of ARGs and gene variants between species, and between microbiome communities, and show us novel mobile elements containing ARGs, which may be potentially transferred horizontally between species. We will apply machine learning approaches to quantitatively study how each bacterial species responds to and survives under antibiotic exposure, by analyzing genomic variants in ARGs, changes in three dimensional structure of the key ARGs, ARG network and pathways.